Statistical Modeling of Extreme Values with Applications to Air Pollution
H. M. Barakat1, E. M. Nigm1 and O. M. Khaled2
1Department of Mathematics Faculty of Science Zagazig University, Zagazig, Egypt
2Department of Basic Science Faculty of Engineering, Sinai University, El-Arish, Egypt Email
Abstract: In this paper the Block Maxima and the Peak Over Threshold methods are used to model the air pollution in two cities in Egypt. A simulation technique is suggested to choose a suitable threshold value. The validity of full bootstrapping technique for improving the estimation parameters in extreme value models has been checked by Kolmogorov-Smirnov test. A new efficiency approach for modeling extreme values is suggested. This approach can convert any ordered data to enlarged block data by using sup-sample bootstrap. Although, this study is applied on three pollutants in two cities in Egypt, but the suggested approaches may be applied on other pollutants in other regions in any country.
[H. M. Barakat, E. M. Nigm and O. M. Khaled. Statistical Modeling of Extreme Values with Applications to Air Pollution. Life Sci J 2012;9(1):124-132] (ISSN: 1097-8135). http://www.lifesciencesite.com. 19
Key words: Air pollution; Generalized extreme value model; Generalized Pareto distribution; Kolmogorov-Smirnov test; Bootstrap technique. Full Text 19